Description and Source of Study Population

Our study population includes several major categories of participants. We included key stakeholders who are knowledgeable about or have programmatic

responsibility for P4HB^® (policy development, outreach/education, enrollment, service provision). In order to capture an accurate description of P4HB^®, we interviewed informants who helped design the program, lobbied for it in the state legislature, as well as those who maintain responsibility for the program’s oversight. Therefore, we

interviewed both state public health and Medicaid staff. Originally, we had hoped to interview key state legislators who were instrumental in providing legislative and economic support in the Georgia legislature. However, these informants could not be recruited, most likely because the legislature was out of session during the recruitment process. We were able, though, to collect rich information from key advocates and DCH officials about the influence of the legislature on the implementation of P4HB^®.

To gain an understanding of how P4HB^® works from the managed care side, we interviewed representatives from each of the three CMOs. These informants had primary knowledge and/or responsibility for the P4HB^® program from their respective

organization. Also, we interviewed several key advocates who represented family

planning and maternal and child health organizations that were instrumental in designing, advocating for, and participating in the early implementation of P4HB^®. To explore eligibility and enrollment processes, we planned originally to interview representatives from PSI, Maximus, and Gammis, all of which have distinct responsibilities in processing

70

P4HB^® applications, as well as assigning P4HB^® enrollees to the CMOs. Instead, we captured this information directly from a review of DCH quarterly reports and interviews with key stakeholders as well as providers. Finally, we interviewed providers who have experience serving the P4HB^® enrolled population. Both public health and private providers were interviewed. Interviews focused on current experience with

outreach/education, enrollment, and service utilization within the P4HB^® program. More detail about the characteristics of all informants is provided in Chapter 4.

We included current and prospective P4HB^® clients in our study sample. These participants were included in one of six focus groups conducted throughout three regions of the state (Atlanta, Southeast Georgia, and Northwest Georgia). The purpose of the focus groups was to assess women’s knowledge of and experience with P4HB^®.

Finally, we conducted a provider survey to gain a more quantitative picture of the P4HB^® program. Approximately104 providers completed a survey that assessed their knowledge and experience with P4HB^®. The survey also addressed whether providers currently serve P4HB^® clients, were provided information or outreach about the program, and what barriers they have experienced with the program. Providers represented

members of three specific provider groups: 1) members of the Georgia Obstetrics and Gynecology Society (GOGS); 2) members of the Georgia Academy of Family Physicians (GAFP), and 3) current Title X providers. Title X, the national family planning program funded by the Office of Population Affairs, is administered in Georgia through the Office of Family Planning in the Department of Public Health.

71 3.6.2. Participant Recruitment

We employed three major strategies for recruiting participants in our study. First, we used “formal networks” to recruit an initial list of key stakeholders. Participants of formal networks use particular services or are part of a common community of interest (Hennick, Hutter et al. 2011). In our case, we recruited participants who were connected in some way to the P4HB^® program. A similar and well-known recruitment strategy is called purposive sampling, whereby participants are recruited based on selected set of criteria, which in this case would be the involvement in and knowledge of P4HB^® (Patton 2008).

We began recruiting participants who were part of Georgia’s state public health and health services community. This list included state DCH (Medicaid) and DPH (Public Health) officials. Due to the researcher’s position at Emory University and past research experience, the research had established relationships with each of these state health agencies. Therefore, this formal network was the most obvious one with which to start recruitment. After introductory emails to prospective informants about the study, initial interviews were scheduled with current Medicaid officials with primary

responsibility and oversight of the P4HB^® program. In addition, a few key interviews with advocates were scheduled based on prior knowledge of their P4HB^® experience as well.

We used a second recruitment strategy known as the snowball method.

“Snowballing” is also known as “chain sampling” whereby participants with specific characteristics who may be difficult to identify or who are unknown to the recruiters can be approached. The snowball strategy involves asking a study participant or key

72

informant whether they know anyone else who meets the study criteria and asking them to refer this person to the researcher (Hennick, al., 2011). We used this snowballing method during our interviews with public health contacts and some advocates to help us connect to providers and provider organizations, such as GOGS, GAFP, and local title X providers whom we interviewed. Our contacts at DCH provided linkages as well to each of the three care management organizations (CMOs) that serve P4HB^® clients

We employed a third and final recruitment strategy, which was the use of gatekeepers. Gatekeepers are people who have a significant and recognized role in the local community and knowledge about the characteristics of community members. In our case, we sought gatekeepers who had: a) knowledge of P4HB^® participants and

providers; and b) access to these participants and can communicate with them on our behalf. The CMOs were one type of gatekeepers that assisted us in recruitment of the focus groups. Each of the CMOs mailed DCH-approved flyers advertising the availability of the focus groups to their current P4HB^® enrollees. One CMO was also instrumental in providing access to one of its Resource Mothers who was interviewed for the study.

Another type of gatekeeper we used was the state Title X program, which facilitated access to its family planning women’s health coordinators and providers. The Title X women’s health coordinators completed the provider survey as well as assisted with introducing us to providers at the local health departments who managed the local family planning programs. With these connections, we were able to establish sites for six of the focus groups, and to conduct four interviews with local Title X providers.

For the interviews, described in more detail below, we recruited study participants and conducted analyses of the information concurrently. Therefore, we stopped

73

recruitment when data saturation was achieved or further recruitment as not possible.

Data saturation in qualitative studies occurs when the researchers no longer “hears” or learns new information (Grady, 2003). Indeed, the saturation of our interview data determined our sample size. However, as noted by Charmaz (2006), oftentimes the sample size is determined by the study aims. Since our aims were relatively

straightforward, whereby we asked informants to describe their involvement and experience with the implementation of P4HB^®, we achieved saturation rather quickly.

Also, as noted in the proposal, we used a stakeholder mapping form to help us keep track of our participant recruitment, (see Appendix B). This form was adapted from the USAID Health Policy Initiative (USAID, 2010). More information about data analysis is

provided below.

Table 3 summarizes data collection for the entire study, which included a total of 21 interviews with 24 informants, six focus groups with 49 participants, surveys from 104 participants, and over 20 document or web-site reviews. The table describes each of the data collection methods by recruitment strategy (where applicable). Following the table is an in-depth description of each data collection method.

74

Table 3: Summary of Data Collection, by Recruitment Strategy

Data Collection Method Recruitment Strategy Number of Participants/Observations Informant interviews Formal Networks

 Medicaid

Focus groups Gatekeepers 6 focus groups with a total of 49 participants

Document reviews Formal networks

 Medicaid

 CMOs

20+ observations including the following:

-P4HB® Concept Paper

-P4HB® marketing plans & budgets -P4HB® outreach/communications plan -P4HB® 2011 and 2012 quarterly reports

-P4HB® website

-Medicaid and CMO P4HB® provider materials

-P4HB® advertising materials (postcards, billboards, flyers)

-P4HB® application materials

75 3.7 Interviews

Twenty-one interviews were conducted with twenty-four informants. We

interviewed four public health officials, two from Georgia’s Department of Community Health (DCH), which administers Medicaid and the P4HB^® program, and two from the Georgia Department of Public Health (DPH), which administers the Maternal and Child Health and Family Planning programs. We also interviewed five advocates who were involved in either the conceptualization, design, and or initial implementation of P4HB^®. These advocates represented family planning, maternal and child health, as well as provider organizations in Georgia. In addition, we interviewed nine interviews with a range of OBGYNs, social workers, resource mothers, and nurses who work in health department family planning programs. We also conducted interviews with representatives from each of the three CMOs.

A semi-structure interview guide was developed to address the major constructs of the theoretical framework and adapted to address the specific type of informant. For instance to address statutory coherence, interview questions for state officials and advocates centered on defining the purpose of the P4HB^® program and discussing the policy planning process. We asked these informants to describe their knowledge of and experience with the conceptualization of P4HB^®. Informants provided details about how the idea of P4HB^® was developed, the motivations for its creation, what stakeholders were involved in the early design phase of the program, as well as the design for long-term goals for P4HB^®. When applicable and as described during the course of some provider interviews, these domains were also addressed. We examined these same issues through a document review of P4HB related materials, including the P4HB^® Concept

76

Paper, CMS STCs, communication plan, provider outreach plan, 2011 quarterly reports, and the 2011 Annual report. Finally, we addressed access to care issues by asking informants to describe certain implementation processes, including enrollment and outreach and education efforts. In most situations, providers and CMO representatives were most knowledge about these issues, along with state public health officials. A copy of the standard interview guide is provided in Appendix C.

The researcher led all interviews and took cursory notes during each interview.

She also created short summaries after each interview. All but one interview was conducted by phone, and the interviews lasted between 30 and 60 minutes, with an average time of 45 minutes. Each interview was tape-recorded with the informant’s permission. Prior to the start of the interview, informants were emailed a copy of the approved informed consent form and asked to return the signed form by fax or email. In some situations, verbal consent was obtained and tape-recorded if the informant did not return a signed consent form. Informants were also provided the opportunity to ask questions about the study prior to the start of the interview. Informants were not compensated for their time, however follow-up communication occurred via email to thank each informant for their time and participation in the study. Informants were also offered the opportunity upon request to obtain a written summary of the interview or the interview transcript. Only one informant made this request after the completion of the interview for both the summary and transcript.

77 3.8 Document review

A document review, also sometimes called “document analysis,” refers to the systematic evaluation of written or electronic materials. Like other qualitative methods, a document review requires analyzing and interpreting data in order to “elicit meaning, gain understanding, and develop empirical knowledge” (Bowen, 2009, page 27).

There are many reasons one might conduct a document review. Most researchers use documents to supplement other sources of data or to corroborate evidence from other sources (Bowen, 2009). Atkinson and Coffey (2004) warn, however, that document reviews should not be used in place of other valuable sources of information, particularly if the information sought is about how an organization or program operates on a day-to-day basis. Instead, documents can provide useful background information and context and should be used in combination with other materials to investigate a particular

research question. Document reviews are particularly useful in case study research when reports and internal correspondence are used as potential sources of empirical data (Mills, Bonner, and Francis, 2006).

For this case study research, we gathered multiple sources of documents to evaluate the implementation of P4HB^®. The primary goal of our document review was to have supplementary materials that provided insight into the conceptualization, policy guidance, and overall goals of the program. A secondary goal of the document review was to understand what the roles and responsibilities were of implementing agencies and providers and how these responsibilities were integrated with Medicaid, which holds jurisdiction over these agencies for P4HB^®. Lastly, the document review allowed us to

78

gather evidence of funding and other resources allocated to the first year of the program’s implementation.

We reviewed over 20 sources of materials. We began chronologically with collecting information that was publicly available about P4HB^®. This included the state appropriations bills for 2008-2010 that reflect the allotment of funding for P4HB in its first year. Next, we reviewed news articles about the program as well as the P4HB^® Concept Paper that was posted online at the Department of Community Health (DCH, 2009). We also reviewed the CMS STCs, which outlines specific responsibilities of DCH for implementing P4HB and collaborating with implementing agencies. Information from this document also reflects CMS expectations about meeting the program’s long-term objectives. Then moving forward, we collected any and all marketing materials such as postcards, flyers, and information posted on the DCH website. The website also

contained program information, such as eligibility criteria, application procedures, and provider information. All website information was included in our analysis. Finally, we collected P4HB^® quarterly reports that were produced in 2011 in the first year of the program. These quarterly reports are required by CMS for all Medicaid demonstration programs. They provided information about certain aspects of the implementation process, including enrollment, outreach, and education, and marketing. We also reviewed internal memos and correspondence that were made available to Ms. Blake through her work on the outcomes evaluation team at Emory. Such memos and correspondence included: marketing budgets and invoices, and internal memos about application and enrollment processes.

79

We developed a Summary Form to review each document included in the study.

This form (see Appendix D) included information about the name, date, and author or source of the document (such and DCH, DPH, CMO). We also noted the category of the documents that were reviewed, including: 1) Policy development; 2) Program guidelines;

3) Marketing/outreach; 4) Quarterly reports; and 5) Provider materials. Policy

development documents included the P4HB® Concept Paper, appropriations bills, and Medicaid statute. Program guidelines included the information gathered from the DCH website about enrollment procedures, CMO selection, and service benefits and coverage.

Marketing and outreach documents included advertisements, flyers, and marketing budgets. Quarterly reports are those also mentioned above that are required by CMS.

Finally, we also reviewed provider guidelines and handbooks available online by each of the CMOs.

The Summary Form was used to capture a general description of each document as well as specific elements of implementation that were pertinent to our study. Such elements included information about 1) The nature of “the problem”; 2) P4HB^® goals and objectives; 3) program design, such as provider network and benefits and services; 4) implementation processes, such as outreach, education, and marketing; 5) program access and utilization, including application and enrollment procedures; and 6) barriers to

implementation. A list of all documents reviewed for this study is provided in Appendix E.

80 3.9 Focus groups

Between June and July 2012, six focus groups were held throughout Georgia with current and prospective P4HB^® clients. Three focus groups were held in two separate public health departments in metropolitan Atlanta, one focus group was held in a public health department in southeast Georgia, and two focus groups were held in two different public health departments in northwest Georgia. These sites were selected in order to provide geographic diversity in the patient population as well as size and location of the public health system.

The overall goal of the focus groups was to learn if P4HB^® was currently reaching its intended audience and how well the program was working. Focus group discussions centered on the following domains: 1) knowledge and understanding of P4HB^® 2) outreach, marketing, and other education efforts concerning P4HB^®: 3) enrollment process and barriers to enrollment; 4) utilization of P4HB^® family planning and IPC services; 5) recommendations for improvements. Participants also completed an anonymous demographic form that asked them to provide information about their race, age, education, insurance status, source and type regular health care services, including family planning, and P4HB® enrollment status. The moderator’s demographic form and discussion guide are provided in Appendix F and G respectively.

As described above, we conducted recruitment in two major ways. First, through our connections with the state Title X family planning program, we sought the assistance of local Title X program managers who helped us identify clinics for each of the six focus groups. The locations were selected based on the following criteria: 1) number of P4HB^® participants seen at clinic; 2) willingness of clinic manager to participate in study and

81

assist with recruitment; 3) available space for focus group. We deemed the first criteria as essential, because this would help facilitate recruitment as well as clinic and client experience with the P4HB^® program. To recruit participants, our first effort included posting flyers around the clinics. The flyers advertised the purpose, time, and place of the focus groups. Ms. Blake’s contact information was also provided on each flyer for women to use if they wanted to sign up or learn more about the focus groups.

Recruitment also occurred through the three CMOs, who mailed flyers to their current P4HB^® enrollees. This flyer also advertised the purpose, date, and time of all focus groups, as provided my contact information as well. All interested persons were screened for eligibility, either by phone or email. Selection criteria for these focus groups include:

1) Age 19-44; 2) current legal U.S. citizen; 3) Georgia resident and 4) English-speaking.

Screening information was also collected about women’s P4HB^® enrollment status (enrolled/not enrolled; FP or IPC patient), and source of regular family planning services.

Originally, we had hoped to hold separate focus groups with P4HB^® enrolled participants according to their FP and IPC status, however, due to the low enrollment of IPC enrollees and women’s uncertainty about their status during the screening process, we decided to include all P4HB^® women in the focus groups.

Focus groups lasted between 42 minutes to one hour, with a mean duration of one hour and 20 minutes. We obtained consent from each participant prior to the focus groups. Ms. Blake moderated each focus group, which was audio recorded. Participants were provided $25 for their time, and refreshments were also provided during each focus group. Additional descriptive information about the focus group participants is provided in Chapter 5.

82 3.10 Provider Surveys

A survey was developed and administered to both prospective and participating P4HB®

providers. The goal of this survey was to assess the knowledge and experiences of

providers. The goal of this survey was to assess the knowledge and experiences of